deadlooooooock

common/common.go

@@ -136,3 +136,55 @@ func MultipartUpload(client *http.Client, url string, path string) (err error) {
 	}
 	return
 }
+
+func MoveFile(sourcePath, destPath string) error {
+	inputFile, err := os.Open(sourcePath)
+	if err != nil {
+		return fmt.Errorf("couldn't open source file: %s", err)
+	}
+	outputFile, err := os.Create(destPath)
+	if err != nil {
+		inputFile.Close()
+		return fmt.Errorf("couldn't open dest file: %s", err)
+	}
+	defer outputFile.Close()
+	_, err = io.Copy(outputFile, inputFile)
+	inputFile.Close()
+	if err != nil {
+		return fmt.Errorf("writing to output file failed: %s", err)
+	}
+	// The copy was successful, so now delete the original file
+	err = os.Remove(sourcePath)
+	if err != nil {
+		return fmt.Errorf("failed removing original file: %s", err)
+	}
+	return nil
+}
+
+type Semaphore interface {
+	Acquire()
+	Release()
+	Close()
+}
+
+type semaphore struct {
+	semC chan struct{}
+}
+
+func NewSemaphore(maxConcurrency int) Semaphore {
+	return &semaphore{
+		semC: make(chan struct{}, maxConcurrency),
+	}
+}
+
+func (s *semaphore) Acquire() {
+	s.semC <- struct{}{}
+}
+
+func (s *semaphore) Release() {
+	<-s.semC
+}
+
+func (s *semaphore) Close() {
+	close(s.semC)
+}

main.go

@@ -3,6 +3,7 @@ package main
 import (
 	"context"
 	"crypto/tls"
+	"flag"
 	"fmt"
 	"log"
 	"net/http"
@@ -25,9 +26,9 @@ import (
 var (
 	arangoDB  adriver.Database
 	arangoCTX context.Context
-	colChunk      adriver.Collection
+	//colChunk      adriver.Collection
 	colFile adriver.Collection
-	colFile2Chunk adriver.Collection
+	//colFile2Chunk adriver.Collection
 	colGMA      adriver.Collection
 	colGMA2File adriver.Collection
 )
@@ -38,7 +39,7 @@ func ConnectDB(baseURL string, arangoUser string, arangoPWD string, arangoDataba
 	// Retry Loop for Failed Connections
 	for i := 0; i < 6; i++ {
 		if i == 5 {
-			return driver, ctx, fmt.Errorf("connectdb: unable to connect to database %d times!", i)
+			return driver, ctx, fmt.Errorf("connectdb unable to connect to database %d times!", i)
 		} else if i > 0 {
 			time.Sleep(30 * time.Second)
 		}
@@ -79,10 +80,10 @@ func ConnectDB(baseURL string, arangoUser string, arangoPWD string, arangoDataba
 func InitDatabase() (err error) {
 	arangoDB, arangoCTX, err = ConnectDB("http://192.168.45.8:8529/", "gma-inator", "gma-inator", "gma-inator")
 
-	colChunk, err = arangoDB.Collection(arangoCTX, "chunk")
+	/*colChunk, err = arangoDB.Collection(arangoCTX, "chunk")
 	if err != nil {
 		return err
-	}
+	}*/
 	colFile, err = arangoDB.Collection(arangoCTX, "file")
 	if err != nil {
 		return err
@@ -91,10 +92,10 @@ func InitDatabase() (err error) {
 	if err != nil {
 		return err
 	}
-	colFile2Chunk, err = arangoDB.Collection(arangoCTX, "file_chunk_map")
+	/*colFile2Chunk, err = arangoDB.Collection(arangoCTX, "file_chunk_map")
 	if err != nil {
 		return err
-	}
+	}*/
 	colGMA2File, err = arangoDB.Collection(arangoCTX, "gma_file_map")
 	if err != nil {
 		return err
@@ -102,8 +103,15 @@ func InitDatabase() (err error) {
 	return nil
 }
 
+var JobPoolSize int = 5
+var ConcurrencyLimit int = 5
+var WorkerJobPool chan string
+
 func main() {
+	folderPathP := flag.String("path", "/mnt/SC9000/TemporaryTestingShit2", "a string")
 	debug.SetMemoryLimit(6e9)
+	flag.Parse()
+	folderPath := *folderPathP
 
 	go func() {
 		log.Println(http.ListenAndServe("0.0.0.0:6060", nil))
@@ -132,14 +140,17 @@ func main() {
 	//fileHandle, err := os.Open("2143898000.1593250551.bin.gma") //2143898000.1593250551.bin")
 	//gma, err := gma.NewReader("2143898000.1593250551.bin.gma")
 
-	folderPath := "/mnt/SC9000/TemporaryTestingShit/"   //"/mnt/worksucc/san1/gma/2/5/4/8/"
+	//folderPath := "/mnt/SC9000/TemporaryTestingShit2/"  //"/mnt/worksucc/san1/gma/2/5/4/8/"
 	folderPathTarget := "/mnt/SC9000/ProcessedGMATest/" //"/mnt/worksucc/san1/gma/2/5/4/8/"
-	//0
+	//
 	entries, err := os.ReadDir(folderPath)
 	if err != nil {
 		panic(err)
 	}
 	skipBla := false
+
+	var WorkerJobPool []string
+
 	for _, e := range entries {
 		if !e.IsDir() && skipBla == true {
 			if e.Name() == "2547463094.1626322945.gma" {
@@ -149,15 +160,35 @@ func main() {
 			}
 		}
 		if !e.IsDir() {
-			err = ProcessGMA(filepath.Join(folderPath, e.Name()))
+			WorkerJobPool = append(WorkerJobPool, filepath.Join(folderPath, e.Name()))
+		}
+	}
+
+	/*
+		sem := common.NewSemaphore(ConcurrencyLimit)
+		wg := sync.WaitGroup{}
+	*/
+	for _, jobFile := range WorkerJobPool {
+		//wg.Add(1)
+		//go func(jobFile string, wg *sync.WaitGroup) {
+		//	sem.Acquire() // Wait for worker to have slot open
+
+		err = ProcessGMA(jobFile)
 		if err != nil {
 			fmt.Printf("\nERROR: %v\n", err)
 			//panic(err)
 			continue
 		}
-			os.Rename(filepath.Join(folderPath, e.Name()), filepath.Join(folderPathTarget, e.Name()))
+		os.Rename(jobFile, filepath.Join(folderPathTarget, filepath.Base(jobFile)))
-		}
+
+		//	sem.Release() // Release the slot
+		//	wg.Done()     // Finish job
+		//}(job, &wg)
 	}
+
+	// Wait for all jobs to finish
+	//wg.Wait()
+
 }
 
 func undoBatch(undoBatch bool, gmaID string, fileIDs []string, gma2FileIDs []string) (err error) {
@@ -339,8 +370,8 @@ func ProcessGMA(filePath string) (err error) {
 		// process and work withj
 		metaSlice, errorSlice, _ := colFile.CreateDocuments(arangoCTX, dboFiles[0:chunkSize])
 
-		//fmt.Println("Metaslice")
+		fmt.Println("Metaslice")
-		//fmt.Println(metaSlice)
+		fmt.Println(metaSlice)
 		for _, meta := range metaSlice {
 			if !meta.ID.IsEmpty() {
 				newUnknownFiles = append(newUnknownFiles, meta.Key)
@@ -394,9 +425,10 @@ func ProcessGMA(filePath string) (err error) {
 		unknownFileID := fmt.Sprintf("file/%s", unknownFile)
 		for _, dboGMA2File := range dboGMA2Files {
 			if unknownFileID == dboGMA2File.File {
-				//fmt.Printf("Uploading %s (local %s) to Storage\n", dboGMA2File.UploadID, dboGMA2File.LocalFileName)
+				fmt.Printf("Uploading %s (local %s) to Storage\n", dboGMA2File.UploadID, dboGMA2File.LocalFileName)
 				err = common.MultipartUpload(httpClient, fmt.Sprintf("http://127.0.0.1:13371/stash/%s/%d", dboGMA2File.UploadID, dboGMA2File.FileSize), dboGMA2File.LocalFileName)
 				if err != nil {
+					fmt.Println("oopsie")
 					undoBatch(true, dboGMA.ID, fileIDs, gma2FileIDs)
 					return err
 				}
@@ -407,6 +439,7 @@ func ProcessGMA(filePath string) (err error) {
 	// TODO : write new gma from arangoinfo
 	// TODO : compare hashes
 	{
+		fmt.Println("rewriting gma")
 		destPath := filepath.Join(gmaTempPath, "rewrite.gma")
 		dir := filepath.Dir(destPath)
 
@@ -504,6 +537,5 @@ func ProcessGMA(filePath string) (err error) {
 		undoBatch(true, dboGMA.ID, fileIDs, gma2FileIDs)
 		return err
 	}
-	0
 	return nil
 }

storageserver/storageserver.go

@@ -26,9 +26,9 @@ import (
 )
 
 var (
-	PoolMaxItems  = 2500
+	PoolMaxItems  = 500
 	PoolPathFinal = "/mnt/SC9000/storagePools"
-	PoolPathTemp  = "/mnt/SC9000/storageTemp"
+	PoolPathTemp  = "/mnt/ramfs/"
 )
 
 type Pool struct {
@@ -58,9 +58,9 @@ type PoolMaster struct {
 	CurrentPool *Pool
 	lock        sync.Mutex
 
-	LocalPools []Pool
+	LocalPools []*Pool
-	FullPools  []Pool
+	FullPools  []*Pool
-	WORMPools  map[string]Pool
+	WORMPools  map[string]*Pool
 }
 type PoolPackResult struct {
 	PoolID         string
@@ -207,7 +207,7 @@ func (p *Pool) Fetch(id string, writer io.Writer) (err error) {
 func NewPoolMaster(finalPath string, cachePath string) (poolMaster PoolMaster, err error) {
 	poolMaster.finalPath = finalPath
 	poolMaster.cachePath = cachePath
-	poolMaster.WORMPools = make(map[string]Pool)
+	poolMaster.WORMPools = make(map[string]*Pool)
 	//poolMaster.lock = sync.Mutex{}
 
 	destPath := filepath.Join(poolMaster.cachePath, "pool")
@@ -228,8 +228,9 @@ func NewPoolMaster(finalPath string, cachePath string) (poolMaster PoolMaster, e
 	}
 	return poolMaster, nil
 }
-func (p *PoolMaster) NewPool() (pool *Pool, err error) {
+func (p *PoolMaster) NewPool() (*Pool, error) {
-	pool = &Pool{}
+	var err error
+	pool := Pool{}
 	pool.PoolID = uuid.NewV4().String()
 	pool.Finalized = false
 	pool.ReadOnly = false
@@ -238,35 +239,39 @@ func (p *PoolMaster) NewPool() (pool *Pool, err error) {
 	//dir := filepath.Dir(destPath)
 	err = os.MkdirAll(destPath, os.ModePerm)
 	if err != nil {
-		return pool, err
+		return &pool, err
 	}
 
-	return pool, nil
+	return &pool, nil
 }
 
 func (p *PoolMaster) GetCurrentWriteablePool() (pool *Pool, err error) {
 	//fmt.Printf("Current Pool %s, ItemCount = %d\n", pool.PoolID, pool.itemCount)
 	if p.CurrentPool != nil && p.CurrentPool.itemCount >= PoolMaxItems {
+		fmt.Printf("Aquiring Lock for GetCurrentWriteablepool\n")
 		p.lock.Lock()
 		defer p.lock.Unlock()
+		defer fmt.Printf("unlock GetCurrentWriteablepool\n")
+		fmt.Printf("Aquired Lock success for GetCurrentWriteablepool\n")
 		p.CurrentPool.ReadOnly = true
-		p.FullPools = append(p.FullPools, *p.CurrentPool)
+		p.FullPools = append(p.FullPools, p.CurrentPool)
 		// queue for compression
-		fmt.Printf("GetCurrentWriteablePool(): current Pool (%s) is full (%d), creating new one", p.CurrentPool.PoolID, p.CurrentPool.itemCount)
+		fmt.Printf("GetCurrentWriteablePool(): current Pool (%s) is full (%d), creating new one\n", p.CurrentPool.PoolID, p.CurrentPool.itemCount)
 		p.CurrentPool = nil
 	}
 	if p.CurrentPool == nil {
-		pool, err = p.AcquireNewOrRecoverPool()
+		fmt.Printf("Creating new Pool")
+		p.CurrentPool, err = p.AcquireNewOrRecoverPool()
+		fmt.Printf("... got [%s]\n", p.CurrentPool.PoolID)
 		if err != nil {
 			return pool, err
 		}
-		p.CurrentPool = pool
-		return pool, nil
 	}
 
 	return p.CurrentPool, nil
 }
-func RestorePoolFromFolder(folderPath string) (pool Pool, err error) {
+func RestorePoolFromFolder(folderPath string) (pool *Pool, err error) {
+	pool = &Pool{}
 	pool.PoolID = path.Base(folderPath)
 
 	entries, err := os.ReadDir(folderPath)
@@ -285,8 +290,11 @@ func RestorePoolFromFolder(folderPath string) (pool Pool, err error) {
 	return pool, err
 }
 func (p *PoolMaster) ScanForLocalPools() (err error) {
+	fmt.Printf("Aquiring Lock for ScanForLocalPools\n")
 	p.lock.Lock()
 	defer p.lock.Unlock()
+	defer fmt.Printf("unlock ScanForLocalPools\n")
+	fmt.Printf("Aquired Lock success for ScanForLocalPools\n")
 	entries, err := os.ReadDir(filepath.Join(p.cachePath, "pool"))
 	if err != nil {
 		return err
@@ -401,7 +409,10 @@ func (p *PoolMaster) MovePoolPackToWORM(packResult PoolPackResult) (err error) {
 	startTime := time.Now()
 
 	targetFileName := filepath.Join(p.finalPath, fmt.Sprintf("%s.tar", packResult.PoolID))
-	os.Rename(packResult.outputFileName, targetFileName)
+	err = common.MoveFile(packResult.outputFileName, targetFileName)
+	if err != nil {
+		return err
+	}
 
 	tarFileCheck, err := os.Open(targetFileName)
 	if err != nil {
@@ -427,8 +438,11 @@ func (p *PoolMaster) PackPool(poolID string) (packResult PoolPackResult, err err
 	startTime := time.Now()
 	packResult.PoolID = poolID
 
+	fmt.Printf("Aquiring Lock for PackPool(%s)\n", poolID)
 	p.lock.Lock()
 	defer p.lock.Unlock()
+	defer fmt.Printf("unlock PackPool\n")
+	fmt.Printf("Aquired Lock success for PackPool(%s)\n", poolID)
 
 	packResult.outputFileName = filepath.Join(p.cachePath, "pool", fmt.Sprintf("%s.tar", poolID))
 	tarFile, err := os.Create(packResult.outputFileName)
@@ -521,8 +535,8 @@ func (p *PoolMaster) PackPool(poolID string) (packResult PoolPackResult, err err
 func (p *PoolMaster) AcquireNewOrRecoverPool() (pool *Pool, err error) {
 	// p.NewPool()
 	for _, localPool := range p.LocalPools {
-		if !localPool.ReadOnly {
+		if !localPool.ReadOnly && localPool.itemCount < 500 {
-			return &localPool, nil
+			return localPool, nil
 		}
 	}
 	return p.NewPool()
@@ -585,8 +599,11 @@ func (p *PoolMaster) FetchLoadWORM(chunkID string, fileID string, writer io.Writ
 	// else load wormPool into disk-cache extract to "worm"
 	// wormMode
 
+	fmt.Printf("Aquiring Lock for FetchLoadWORM\n")
 	p.lock.Lock()
 	defer p.lock.Unlock()
+	defer fmt.Printf("unlock FetchLoadWORM\n")
+	fmt.Printf("Aquired Lock success for FetchLoadWORM\n")
 
 	var dboChunk common.DB_Chunk
 	_, err = colChunk.ReadDocument(arangoCTX, chunkID, &dboChunk)
@@ -607,7 +624,7 @@ func (p *PoolMaster) FetchLoadWORM(chunkID string, fileID string, writer io.Writ
 		return err
 	}
 	fmt.Println("extracted")
-	p.WORMPools[loadedWormPool.PoolID] = loadedWormPool
+	p.WORMPools[loadedWormPool.PoolID] = &loadedWormPool
 	return loadedWormPool.Fetch(fileID, writer)
 	//return nil
 }
@@ -707,9 +724,11 @@ func (p *PoolMaster) Store(id string, src io.Reader, targetSize int64) (err erro
 	}
 
 	fmt.Printf("Store(%s)\n", id)
-
+	fmt.Printf("Aquiring Lock for Store\n")
 	p.lock.Lock()
 	defer p.lock.Unlock()
+	defer fmt.Printf("unlock Store\n")
+	fmt.Printf("Aquired Lock success for Store\n")
 	// figuring out paths
 	poolFolder := filepath.Join(p.cachePath, "pool", pool.PoolID)
 	destPath := filepath.Join(poolFolder, id)
@@ -764,7 +783,9 @@ func (p *PoolMaster) Store(id string, src io.Reader, targetSize int64) (err erro
 
 	return nil
 }
-
+func removeFromSlice(slice []*Pool, s int) []*Pool {
+	return append(slice[:s], slice[s+1:]...)
+}
 func main() {
 	err := InitDatabase()
 	if err != nil {
@@ -781,8 +802,39 @@ func main() {
 		panic(err)
 	}
 
+	go func() {
+		for {
+			for index, fullPool := range poolMaster.FullPools {
+				poolMaster.lock.Lock()
+
+				packResult, err := poolMaster.PackPool(fullPool.PoolID)
+				if err != nil {
+					panic(err)
+				}
+				err = poolMaster.ImportPoolPackResult(packResult)
+				if err != nil {
+					panic(err)
+				}
+				err = poolMaster.MovePoolPackToWORM(packResult)
+				if err != nil {
+					panic(err)
+				}
+				os.RemoveAll(filepath.Join(poolMaster.cachePath, "pool", fullPool.PoolID))
+				poolMaster.FullPools = removeFromSlice(poolMaster.FullPools, index)
+
+				poolMaster.lock.Unlock()
+			}
+			time.Sleep(time.Second * 10)
+		}
+	}()
 	for _, localPool := range poolMaster.LocalPools {
 		if localPool.ReadOnly {
+
+			dboChunkExists, err := colChunk.DocumentExists(arangoCTX, localPool.PoolID)
+			if err != nil {
+				panic(err)
+			}
+			if !dboChunkExists {
 				fmt.Printf("Packing Pool %s\n", localPool.PoolID)
 				packResult, err := poolMaster.PackPool(localPool.PoolID)
 				if err != nil {
@@ -796,6 +848,9 @@ func main() {
 				if err != nil {
 					panic(err)
 				}
+				os.RemoveAll(filepath.Join(poolMaster.cachePath, "pool", localPool.PoolID))
+			}
+			//os.RemoveAll(filepath.Join(poolMaster.cachePath, "pool", localPool.PoolID))
 		}
 		// packResult.FileCount
 	}